Chapter II. Achieving Safety Outcomes

This chapter presents the findings of the analyses of data from the National Child Abuse and Neglect Data System (NCANDS) and the Adoption and Foster Care Analysis and Reporting System (AFCARS) pertaining to both child maltreatment population characteristics and State performance on outcomes pertaining to child safety. The number of States included in each analysis varies because of differences in the number of States reporting data for a particular variable or outcome measure.

The chapter includes a description of: (1) the characteristics of State child maltreatment populations, and (2) State performance on national measures of child safety. State performance is reported with respect to the data for 2002, and change in performance is reported based on data from 1999 to 2002. A State was determined to exhibit a change in performance from 1999 to 2002 if the percent change was greater than 5.0 in either direction.¹⁰

This chapter also includes the findings regarding the relationships between the characteristics of State maltreatment populations and State performance on child safety outcome measures. The initial decision to include information pertaining to the characteristics of State child maltreatment populations in the Report to Congress State data pages was based on the assumption that a comprehensive understanding of State performance with regard to achieving the national outcomes may require a consideration of contextual factors, such as population characteristics.

The relationships between population characteristics and State outcome performance were assessed using Spearman’s rank order correlation analysis (rho). Because the data represent the entire population of maltreatment reports rather than a sample of those reports, the determination of statistical significance is not appropriate. Instead, Guilford’s suggested interpretation for correlation values, provided in Appendix G, is used to identify the strength of the relationship.¹¹

Characteristics of State Maltreatment Populations

According to the NCANDS data, in 2002, the rate of child victims¹² per 1,000 children in the population varied considerably across States, from 1.8 child victims per 1,000 in the population in Pennsylvania¹³ to 41.3 child victims per 1,000 in the population in Alaska, with a median of 10.8 child victims. Six States had child victim rates of less than 5 per 1,000 children in the population (Arizona, New Hampshire, New Jersey, Pennsylvania, Virginia, and Washington). In comparison, 5 States had victim rates of more than 20 per 1,000 children in the population (Alaska, the District of Columbia, Florida, Massachusetts, and South Dakota).

Analyses of NCANDS data indicate a similar variation across States with regard to substantiation rates. The substantiation rate, as distinguished from the rate of child victims per 1,000 children in the population, addresses how likely it is that a State will substantiate a report of child maltreatment. Thus, States could have the same number of reports of maltreatment, yet one State may substantiate a larger number of reports. Consequently, the substantiation rate may be defined as the number of children who were determined to be victims of a “substantiated” or “indicated” maltreatment report, divided by the total number of children who were the subject of an investigated report alleging child maltreatment. In 2002, State substantiation rates ranged from 9.1 percent in New Hampshire to 79.5 percent in Alaska, with a median of 27.5 percent. Three States had substantiation rates of at least 50 percent (Alaska, Hawaii, and Massachusetts). In contrast, 5 States had substantiation rates of 15 percent or less (Arizona, Missouri, Montana, New Hampshire, and New Jersey).

Neither the substantiation rate nor the rate of child victims in 2002 was found to be related to the size of a State’s child population. However, a strong relationship was found between the child victim rate and the substantiation rate (Spearman’s rho = +0.711). A State with a relatively high substantiation rate tended to have a relatively high rate of child victims per 1,000 children in the population, while a State with a relatively low substantiation rate tended to have a relatively low rate of child victims per 1,000 children in the population. This finding suggests that the rate of child victims in a State is in part a function of State policies and definitions regarding child maltreatment

State variations with regard to the age of child maltreatment victims

The information provided in table II-1 below indicates that in 2002, there was considerable variation among States with respect to the age of child victims. In most States, the majority of child victims were younger than age 6. However, in 50 percent of the States, at least 30 percent of child victims were 11 years of age or older. This suggests that the common perception of child maltreatment victims as being very young children is not accurate in many States.

Table II-1. Range and median of State percentages of child victims in 2002 by age group (N=51)

State variation with regard to the race/ethnicity of child maltreatment victims

As shown in table II-2 below, States vary considerably with regard to the race/ethnicity of child victims. In 2002 the majority of child victims in most States were White.¹⁴ There were only 5 States in which 15 percent or more of child victims were Alaska Native/Native American (Alaska, Montana, North Dakota, Oklahoma, and South Dakota). In addition, although Asian/Pacific Islander children comprised 35.3 percent of the child victim population in Hawaii, the next highest percentage was 3.4 in California.

Table II-2. Range and median of State percentages of child victims in 2002 by race/ethnicity (N=50)

Table II-3 provides information pertaining to the relationship between the percentage of children in the State population who were of a particular race/ethnicity and the percentage of that State’s child victims who were of the same race/ethnicity.

Table II-3. Relationship between the percentage of child victims by race/ethnicity and the percentage of State child population by race/ethnicity (All 2002 numbers count at the Report-Child pair level so the same child may be reported more than once in these data) ^*

^*The information is presented for the number of States with the percentage of States shown in parentheses. back

The following are the key findings relevant to this table.

• There were no States in which the percentage of White child victims exceeded the percentage of White children in the population.
• In 6 States, the percentage of Hispanic child victims exceeded the percentage of Hispanic children in the population (Colorado, Connecticut, Massachusetts, Minnesota, North Dakota, and Utah).
• In 32 States, the percentage of Black child victims exceeded the percentage of Black children in the population. In 22 of these 32 States, the percentage of Black child victims was at least twice the percentage of Black children in the State population; in one State (Minnesota), the percentage of Black child victims was five times the percentage of Black children in the population.
• In 13 States, the percentage of Alaska Native/Native American child victims exceeded the percentage of these children in the population. In 10 of these 13 States, the percentage of Alaska Native/Native American child victims was at least twice the percentage in the population; in 7 of the 13 States the percentage of Alaska Native/Native American child victims was three times the percentage in the population (Idaho, Minnesota, Nebraska, Oregon, South Dakota, Washington, and Wisconsin).

State variation with regard to types of maltreatment experienced by child victims

As shown in table II-4, in 2002, the types of maltreatment experienced by child victims varied across States. Although in most States, the majority of child victims were victims of “neglect (not including medical neglect),” there were a few States in which this type of maltreatment accounted for only a small percentage of child victims.

Some of the variation depicted in table II-4 may be attributed to differences in State definitions of maltreatment. For example, in some States, psychological/emotional abuse is incorporated into the State definition of child abuse and neglect, while in other States this type of abuse generally is not part of the legal definition of child maltreatment. As another example, in Pennsylvania, neglect is handled by the Temporary Assistance to Needy Families agency and the State does not report allegations of neglect to NCANDS unless it is determined to be “severe neglect.” Consequently, for the NCANDS data, Pennsylvania had the lowest percentage of children reported as victims of neglect. Furthermore, because Pennsylvania does not include most neglect in NCANDS reporting, other categories of abuse, such as sexual abuse, appear magnified.

Table II-4. Range and median of State percentages of child victims in 2002 by types of maltreatment

No meaningful relationship was found between the types of maltreatment experienced by child victims and the ages or race/ethnicity of child victims.

Return to Table of Contents

State Performance With Regard to Keeping Children Safe

• Outcome 1—Reduce recurrence of child abuse and/or neglect.
     Outcome measure 1.1—Of all children who were victims of substantiated or indicated child abuse and/or neglect during the first 6 months of the reporting period, what percentage had another substantiated or indicated report within a 6-month period?
• Outcome 2—Reduce the incidence of child abuse and/or neglect in foster care.
     Outcome measure 2.1—Of all children who were in foster care during the reporting period, what percentage was the subject of substantiated or indicated maltreatment by a foster parent or facility staff member?

Key findings with regard to State performance on these outcome measures for 2002 and changes in State performance from 1999 to 2002 are presented below.

State performance regarding maltreatment recurrence

In 2002, 42 States provided sufficient data to NCANDS to calculate the Report to Congress measure of maltreatment recurrence. Analyses of these data resulted in the following findings:

• The percentage of maltreatment recurrence within a 6-month period ranged from 1.2 in Delaware to 13.7 in New York, with a median of 7.5.
• Sixteen States (38 percent) met the national standard of 6.1 percent or less for the measure of maltreatment recurrence (Arkansas, Arizona, Colorado, Delaware, Hawaii, Idaho, Maine, Minnesota, Mississippi, Nebraska, New Hampshire, Pennsylvania, South Carolina, Texas, Vermont, and Virginia).

Information pertaining to change in performance on this measure from 1999 to 2002 is provided in table II-5. As shown in the table, State performance exhibited an increase from 1999 to 2002 in the percentage of States meeting the national standard (+11.8 percent change). However, the increase in the number of States reporting from 1999 to 2002 suggests that these findings should be interpreted with caution, because different States were represented in the population from one year to the next.

Table II-5.State performance from 1999 through 2002 with regard to maltreatment recurrence (outcome measure 1.1)

Figure II-1 shows the percentages of States exhibiting improvement, decline, or no change in performance from 1999 to 2002 for the States that provided data in both years. A State was determined to exhibit a change in performance from 1999 to 2002 if the percent change was greater than plus or minus 5.0.¹⁵ Two States that improved in performance from 1999 to 2002 met the national standard in both time periods (i.e., 1999 and 2002). Two States (Hawaii and Vermont) went from not meeting the national standard in 1999 to meeting the national standard in 2002.

State performance regarding maltreatment in foster care

In FY 2002, 42 States provided sufficient data to NCANDS and AFCARS to calculate outcome measure 2.1—the percentage of children in foster care for the first 9 months of the calendar year who were maltreated by foster parents or facility staff. Analyses of these data resulted in the following findings:

• The percentage of children in foster care who were victims of maltreatment by foster parents or facility staff ranged from 0.05 in Vermont to 1.62 in Oklahoma, with a median of 0.39.
• Twenty-five (59 percent) of the 42 States met the national standard of 0.57 or less for this measure.

Table II-6 provides data pertaining to State performance on this measure from 1999 through 2002. As shown in the table, the number of States providing sufficient data to calculate the measure doubled from 1999 to 2002. Both the mean performance and median performance exhibited improvement from 1999 to 2002. Although the percentage of States meeting the national standard declined slightly from 2001 to 2002, the percentage in 2002 represents a +13.8 percent change from 1999. Again, the increase in the number of States reporting from 1999 (21 States) to 2002 (42 States) suggests that these findings should be interpreted with caution, because different States are represented in the population from one year to the next.

Table II-6. State performance from 1999 to 2002 with regard to maltreatment of children in foster care by foster parents or facility staff (outcome measure 2.1)

Figure II-2 presents the percentage of States exhibiting improvement, decline, or no change in performance from 1999 to 2002 (for the 21 States that provided data in both years). Five of the 10 States that improved in performance from 1999 to 2002 met the national standard in both 1999 and 2002. Four of the 10 States that declined in performance from 1999 to 2002 continued to meet the national standard despite the decline. Only one State (Florida) went from not meeting the standard in 1999 to meeting the standard in 2002.

Return to Table of Contents

State Child Victim Characteristics and State Performance With Regard to Maltreatment Recurrence

Analyses were conducted to assess whether various aspects of the State child welfare context affected performance on outcome measure 1.1. The contextual factors examined were the following:

• Rate of child victims per 1,000 children in the population.
• Age of child victims.
• Race/ethnicity of child victims.
• Type of maltreatment experienced by child victims.

A substantial positive relationship (according to Guildford’s suggested interpretation provided in Appendix G) was found between the rate of child victims per 1,000 children in the population and State performance on outcome measure 1.1 (Spearman’s rho = .613). In general, States with a relatively high rate of child victims tended to have a relatively high rate of maltreatment recurrence within a 6-month period. In contrast, States with a relatively low rate of child victims tended to have a relatively low rate of maltreatment recurrence within a 6-month period.

The following are specific findings with regard to this relationship:

• Thirteen of the 21 States with child victim rates that were less than the national median met the national standard of 6.1 percent or less for the maltreatment recurrence measure.
• Only 2 of the 21 States with victim rates greater than the national median met the national standard of 6.1 percent or less.
• Sixteen of the 20 States that exceeded the national median for the measure of maltreatment recurrence also exceeded the national median for the rate of child victims in 2002.

One explanation for the relationship may be that State definitions of maltreatment that are relatively broad in scope compared to those of other States may result in both higher victim rates and higher maltreatment recurrence rates as calculated by outcome measure 1.1. This suggests that State policies and definitions regarding child maltreatment may impact both the number of victims and the incidence of maltreatment recurrence within a 6-month period.

A negative substantial relationship was found between the percentage of child victims experiencing sexual abuse and State performance on the measure of maltreatment recurrence (Spearman’s rho = -0.489). States with a relatively high percentage of child victims experiencing sexual abuse tended to have a relatively low incidence of maltreatment recurrence. Correspondingly, States with a relatively low percentage of child victims experiencing sexual abuse tended to have a relatively high incidence of maltreatment recurrence. It is difficult to interpret this finding without additional research information pertaining to the relationship between maltreatment recurrence and child sexual abuse.

No substantial relationships were found between the percentage of children experiencing psychological or emotional abuse and State performance on the measure of maltreatment recurrence. Also, no substantial relationships were found between the percentage of child victims who were of a particular age or race/ethnicity and State performance on the outcome measure of maltreatment recurrence.

¹⁰Percent change is calculated by subtracting the percentage for 1999 from the percentage for 2002 and dividing by the percentage for 1999. Back

¹¹Guilford, J.P. (1956). Fundamental statistics in psychology and education (3rd ed.). New York: McGraw Hill. Back

¹²A child victim is defined as the subject of a substantiated or indicated maltreatment report. Back

¹³Pennsylvania’s victim rate is due to the fact that the State does not report child neglect to NCANDS unless it is severe neglect. Back

¹⁴For purposes of this report, the term White refers to all children identified as White who are not Hispanic. The term Black refers to all children identified as Black who are not Hispanic. The term Hispanic refers to all children identified as being of Hispanic origin, regardless of race. Back

¹⁵Percent change is calculated by subtracting the percentage for 1999 from the percentage for 2002 and dividing by the percentage for 1999. Back

Return to Table of Contents

Download FREE Adobe Acrobat® Reader™ to view PDF files located on this site.

Children's Bureau Home | About the Children's Bureau | Programs & Funding | Child Welfare Monitoring | Training & Technical Assistance | Frequently Requested Information | Laws & Policies | Statistics & Research | Federal & State Reporting Systems | Upcoming Events | Current Initiatives & Issues | New on Site | Children's Bureau Express Online Digest | Sitemap

ACF Home | Questions? | Site Index | Contact Us | Accessibility | Privacy Policy | Freedom of Information Act | Disclaimers
Department of Health and Human Services | USA.gov: The U.S. government's official web portal

Administration for Children and Families • 370 L'Enfant Promenade, S.W. • Washington, D.C. 20447

Last Updated: May 10, 2006